Progressing cancers lose their differentiation, and acquire multiple genomic aberrations and treatment resistance, what can be enhanced by tumor necrosis factor (TNF). Yet, T cell-based cancer immune therapies, require interferon (IFN)-γ and TNF producing type I T cells, to induce cancer regression and stable remission from metastatic disease. We therefore analyzed the effect of IFN-γ and TNF producing type I T cells on the phenotypic and genomic cancer cell differentiation, in endogenously developing islet cell cancers of RT2 mice. During progression, islet cancers progressively lose first the glucose transporter 2, then insulin and finally synaptophysin, and acquire multiple genomic aberrations. Immune therapy with type I T cells and immune checkpoint inhibitors (ICI), prevented the genomic aberrations and promoted differentiation. This required the joint signaling of IFN-γ and TNF, as type I T cells and ICI therapy prevented neither antigen loss nor the acquisition of genomic aberrations in either STAT1- or TNF-receptor1-deficient mice. STAT1-expression was also required to prevent metastatic disease. As cancer cell differentiation required the IFN-γ-dependent induction of p16, IFN-γ and TNF from type I immune responses promote genomic stabilization and the p16-dependent differentiation, in developing and established cancers.

Tumor control via immune responses is dependent on sufficient immigration of immune cells into the microenvironment. In addition to the mechanism of tumor cell killing, immune responses can also control tumor cells via soluble factors, like cytokines IFN- and TNF. Tumor control by IFN-γ and TNF induces specific senescence markers in tumor cells and has thus been coined as cytokine-induced senescence (CIS).In vitro, using primary melanoma cells from patients, we demonstrated that not only the combined action of IFN-γ signaling via STAT1 and TNF, but also the combined action of IFN-α signaling via STAT1 and STAT3 and TNF induces CIS in melanoma cells. We therefore treated two patients who developed ascites due to metastatic melanoma after all therapies had failed with intraperitoneal IFN-α as the ascites fluid contains endogens TNF. IFN-α therapy was associated with a strong reduction in ascites and melanoma cells. After IFN-α therapy, melanoma cells expressed either the cell cycle inhibitors p16 or nuclear phospho-p21. Moreover, they remained growth arrested when cultured in vitro, whereas melanoma cells isolated before therapy proliferated strongly. To experimentally investigate whether CIS can control ascites, either IFN-α- and TNF-resistant WM115 tumor cells or IFN-α- and TNF-sensitive A204 tumor cells were injected intraperitoneally into mice. IFN-α therapy of the respective mice controlled the growth of intraperitoneal A204 cells only, whereas intraperitoneal WM115 cells grew exponentially even in the presence of IFN-α and TNF. Overall, the data show that the combined effect of IFN and TNF contributes to the control of metastatic cancer cells, but only when the IFN-induced senescence signaling pathway is present.

Immune checkpoint blockade (ICB) therapy is a central pillar of melanoma treatment leading to durable response rates. Important mechanisms of action of ICB therapy include disinhibition of CD4+ and CD8+ T cells. Stimulated CD4+ T helper 1 cells secrete the effector cytokines interferon-gamma (IFN-γ) and tumor necrosis factor alpha (TNF), which induce senescence in tumor cells. Besides being growth-arrested, senescent cells are metabolically active and secrete a large spectrum of factors, which are summarized as senescence-associated secretory phenotype (SASP). This secretome affects the tumor growth. Here, we compared the SASP of cytokine-induced senescent (CIS) cells with the SASP of therapy-induced senescent (TIS) cells. Therefore, we established in vitro models for CIS and TIS in melanoma. The human melanoma cell lines SK-MEL-28 and WM115 were treated with the cytokines IFN-γ and TNF as CIS, the chemotherapeutic agent doxorubicin, and the cell cycle inhibitor palbociclib as TIS. Then, we determined several senescence markers, i.e., growth arrest, p21 expression, and senescence-associated β-galactosidase (SA-β-gal) activity. For SASP analyses, we measured the regulation and secretion of several common SASP factors using qPCR arrays, protein arrays, and ELISA. Each treatment initiated a stable growth arrest, enhanced SA-β-gal activity, and-except palbociclib-increased the expression of p21. mRNA and protein analyses revealed that gene expression and secretion of SASP factors were severalfold stronger in CIS than in TIS. Finally, we showed that treatment with the conditioned media (CM) derived from cytokine- and palbociclib-treated cells induced senescence characteristics in melanoma cells. Thus, we conclude that senescence induction via cytokines may lead to self-sustaining senescence surveillance of melanoma.

Melanomas are malignant tumors that can partly and very rarely completely regress in response to immune responses. Analyzing the mechanisms underlying this immune-mediated rejection, melanomas became leading in developing general cancer immunotherapy. This resulted in the discovery of tumor-specific neoantigens and mutations autoantigens, now called tumor-associated antigens, and their specific recognition by cytotoxic T lymphocytes. Melanomas were of key importance for the development of adoptive T-cell therapy and active tumor vaccines, namely dendritic cell vaccines. Melanoma therapy with antibodies against CTLA-4 provided the proof of concept that solid cancers can be susceptible to cancer immunotherapy, and melanoma therapy with antibodies against PD-1 resulted in the clinical breakthrough of cancer immunotherapy. Still, about half of patients die from metastatic melanoma. Combining anti‒PD-1 with anti‒CTLA-4 antibodies to increase antitumor immune responses or with targeted therapy improves the overall survival only partially. Recent data revealed a close link between defects in the IFN-γ‒dependent induction of cell cycle control genes and resistance to immunotherapy, which may allow for identifying those patients that respond to immunotherapy and to develop novel therapies, combining cancer immunotherapy with cell cycle inhibitors.

Background Although antibodies blocking immune checkpoints have already been approved for clinical cancer treatment, the mechanisms involved are not yet completely elucidated. Here we used a λ-MYC transgenic model of endogenously growing B-cell lymphoma to analyze the requirements for effective therapy with immune checkpoint inhibitors.Methods Growth of spontaneous lymphoma was monitored in mice that received antibodies targeting programmed cell death protein 1 and cytotoxic T lymphocyte-associated protein-4, and the role of different immune cell compartments and cytokines was studied by in vivo depletion experiments. Activation of T and natural killer cells and the induction of tumor senescence were analyzed by flow cytometry.Results On immune checkpoint blockade, visible lymphomas developed at later time points than in untreated controls, indicating an enhanced tumor control. Importantly, 20% to 30% of mice were even long-term protected and did never develop clinical signs of tumor growth. The therapeutic effect was dependent on cytokine-induced senescence in malignant B cells. The proinflammatory cytokines interferon-γ (IFN-γ) and tumor necrosis factor (TNF) were necessary for the survival benefit as well as for senescence induction in the λ-MYC model. Antibody therapy improved T-cell functions such as cytokine production, and long-time survivors were only observed in the presence of T cells. Yet, NK cells also had a pronounced effect on therapy-induced delay of tumor growth. Antibody treatment enhanced numbers, proliferation and IFN-γ expression of NK cells in developing tumors. The therapeutic effect was fully abrogated only after depletion of both, T cells and NK cells, or after ablation of either IFN-γ or TNF.Conclusions Tumor cell senescence may explain why patients responding to immune checkpoint blockade frequently show stable growth arrest of tumors rather than complete tumor regression. In the lymphoma model studied, successful therapy required both, tumor-directed T-cell responses and NK cells, which control, at least partly, tumor development through cytokine-induced tumor senescence.

Immune checkpoint blockade is one of the central pillars of therapy for metastatic melanoma. Recently, cell cycle inhibitors such as palbociclib have demonstrated antitumor activity in vivo in different tumour entities. Besides tumour cell killing, tumour therapy also induces senescence, which is characterized by a stable growth arrest. The secretome of senescent cells, termed senescence- associated secretory phenotype (SASP), has diverse effects on tumor growth. We wondered how the SASP of therapy-induced senescent (TIS) cells differs from the SASP of cytokine-induced senescent (CIS) cells. Therefore, we induced senescence in two melanoma cell lines using (i) the T helper 1 cell cytokines interferon-gamma and tumornecrosisfactor, (ii) the chemotherapeutic agent doxorubicin, and (iii) palbociclib to compare the different secretomes. To validate senescence induction, we performed different cell cycle analyses with Western Blot and FACS and determined the activity of the senescence-associated β-galactosidase (SA-β-gal). For SASP analyses, we measured the regulation and secretion of several common SASP factors using qPCR arrays, proteome profiler arrays and ELISA. Most measurements were performed at the time of induction and 48 h later to portray the maintenance of senescence. Each treatment initiated a stable growth arrest, enhanced SA-β-gal activity, diminished the proportion of cells in the S phase and, except palbociclib, induced increased expression of p21. PCR array analyses of SASP factors revealed that gene expression in TIS was manifold weaker than in CIS. The protein array analyses confirmed that TIS caused a much less pronounced release of several inflammation- and tumour-promotion-associated factors as compared to CIS. Thus, we conclude that senescence induction via palbociclib could be a promising strategy in the treatment of malignant melanoma, as this form of TIS exhibits a SASP profile with significantly reduced levels of inflammation-associated factors.

Immune checkpoint blockade (ICB)-based or natural cancer immune responses largely eliminate tumours. Yet, they require additional mechanisms to arrest those cancer cells that are not rejected. Cytokine-induced senescence (CIS) can stably arrest cancer cells, suggesting that interferon-dependent induction of senescence-inducing cell cycle regulators is needed to control those cancer cells that escape from killing. Here we report in two different cancers sensitive to T cell-mediated rejection, that deletion of the senescence-inducing cell cycle regulators p16Ink4a/p19Arf (Cdkn2a) or p21Cip1 (Cdkn1a) in the tumour cells abrogates both the natural and the ICB-induced cancer immune control. Also in humans, melanoma metastases that progressed rapidly during ICB have losses of senescence-inducing genes and amplifications of senescence inhibitors. Metastatic cells also resist CIS. Such genetic and functional alterations are infrequent in metastatic melanomas regressing during ICB. Thus, activation of tumour-intrinsic, senescence-inducing cell cycle regulators is required to stably arrest cancer cells that escape from eradication.

Immunotherapy with monoclonal antibodies against exhaustion-associated surface molecules, immune checkpoint blockade (ICB), reactivates T cells and induces durable therapeutic stability in a variety of metastatic cancers. Cancer regression with ICB requires cancer cell killing but additional, unknown mechanisms are needed to establish the long-lasting tumor dormancy in the responder patients. To analyze cancer cell dormancy induced by ICB we used a model of adoptive transfer of tumor-associated antigen (TAA)-specific T-helper-1 (Th1) cells in transgenic tumor-bearing mice. Th1 immunotherapy prolongs the life of mice by induction tumor cell senescence. Combining adoptive transfer of Th1 cells with ICB significantly increased life time. The therapy restored a normal health status in the mice, destroyed the cancers partly and induced a p16Ink4a+/Ki67- senescent phenotype in the remaining cancer cells. The therapy was dependent on an intact interferon Stat1-signaling pathway and the activation of p16Ink4a in the cancer cells. While TAA-specific Th1 cells similarly migrated into cancers of either wildtype or Stat1.ko cancers, they failed to induce p16Ink4a+/Ki67- senescent tumor cells. Stat1.ko and p16Ink4a.ko cancer cells were fully susceptible to apoptosis and T cell-mediated killing. In contrast, both, Stat1.ko and p16Ink4a.ko cancer were resistant to senescence induction, in vitro and in vivo. To determine whether this is also valid for ICB therapy of human melanomas, we analyzed metastatic melanomas that progressed within < 3 months during ICB. These metastasis had losses of senescence-inducing genes or ≥ 4fold amplifications of senescence inhibitors. In contrast such mutations were infrequent in metastatic melanomas that regressed during ICB for > 1 year. Thus, cancer immune control with ICB requires, beside cancer cell killing, the interferon- and p16Ink4a- dependent senescence induction in cancer cells.

Different animal models as well as histological studies using samples from cancer patients clearly showed that cellular senescence is a barrier against tumor progression. Yet, recent data question this concept suggesting that senescent cells may also promote tumor progression. By adopting a senescence-associated secretory phenotype, senescence signaling may even induce dedifferentiation and stemness. To investigate effects of cytokine-induced senescence (CIS) on tumor progression and differentiation, we used pre-malignant β-cancers from RIP-Tag2 mice that consecutively express three differentiation antigens: first synaptophysin, then insulin and last glucose transporter-2 (Glut-2). We show that T-helper-cell-mediated immunity prevented the transition of pre-cancerous adenomas into β-cancers by stabilizing their genome and β-cell functions. Thus, CIS prevented major genomic aberrations, i.e. chromosomal amplifications and losses, by a Stat1- and TNFR1-dependent mechanism. Tumor progression further caused a gradual loss of the functionally relevant differentiation markers insulin and Glut-2. CIS prevented this loss in vivo, again in a strictly cytokine signaling-dependent manner, and preserved the physiological islet functions of the pre-malignant cells ex vivo. Additional experiments revealed that treatment of isolated β-cancer cells with interferon-gamma and tumor necrosis factor retained the expression of β-cell differentiation markers also in vitro. Taken together, immunity-induced senescence surveillance protects against dedifferentiation of endocrine β-cancers thereby enabling the diseased mice to control glucose homeostasis.

Immune checkpoint blockade (ICB) targets exhaustion-associated surface molecules using monoclonal antibodies (mAbs), reactivates T cells and induces therapeutic stability in a variety of metastatic cancers. Cancer regression with ICB involves cancer cell killing but additional, unknown mechanisms are needed to establish long-lasting tumor control in the responder patients. To analyze whether senescence is required for an efficient ICB we used a model of adoptive transfer of tumor-associated antigen (TAA)-specific T-helper-1 (TH1) cells in transgenic tumor-bearing mice. TH1 cell-mediated immunotherapy prolongs the life of mice, combining adoptive transfer of TH1 cells with ICB (anti-PD-L1 and anti-LAG-3 mAbs) significantly increased life time, even in mice with pre-final cancer disease. The therapy restored a normal health status in the mice, partly destroyed the cancers, and induced a p16Ink4a+ and Ki67- senescent phenotype in the remaining cancer cells. The therapy was strictly dependent on an intact interferon/Stat1-signaling pathway, and on the Stat1-mediated activation of p16Ink4a in the cancer cells. Tag2-driven cancers of Stat1.ko mice did not respond to the therapy. While TAA-specific TH1 cells migrated into cancers of either wildtype or Stat1.ko cancers in comparable numbers, they failed to induce senescent tumor cells. Stat1.ko and p16Ink4a.ko cancer cells were fully susceptible to apoptosis and T cell-mediated killing. In contrast, Stat1.ko and p16Ink4a.ko cancer were resistant to senescence induction, in vitro and in vivo. To determine whether this is valid for ICB therapy of patients with metastatic melanomas, we analyzed melanomas that progressed within < 3 months during ICB and melanoma that regressed during ICB for > 1 year. Progressing metastases had losses of senescence-inducing genes or ≥ 4fold amplifications of senescence inhibitors. In contrast, such mutations were infrequent in melanoma metastases that respond to ICB. Thus, therapy with ICB requires induction of cancer cell senescence.

Background/Aims: Cellular senescence, or permanent growth arrest, is known as an effective tumor suppressor mechanism that can be induced by different stressors, such as oncogenes, chemotherapeutics or cytokine cocktails. Previous studies demonstrated that the growth-repressing state of oncogene-induced senescent cells depends on argonaute protein 2 (Ago2)-mediated transcriptional gene silencing and Ago2/Rb corepression of E2F-dependent cell cycle genes. Cytokine-induced senescence (CIS) likewise depends on activation of the p16Ink4a/Rb pathway, and consecutive inactivation of the E2F family of transcription factors. In the present study, we therefore analyzed the role of Ago2 in CIS. Methods: Human cancer cell lines were treated with interferon-gamma (IFN-γ) and tumor necrosis factor (TNF) to induce senescence. Senescence was determined by growth assays and measurement of senescence-associated β-galactosidase (SA-β-gal) activity, Ago2 translocation by Ago2/ Ki67 immunofluorescence staining and western blot analysis, and gene transcription by quantitative polymerase chain reaction (qPCR). Results: IFN-γ and TNF permanently stopped cell proliferation and time-dependently increased SA-β-gal activity. After 24 – 48 h of cytokine treatment, Ago2 translocated from the cytoplasm into the nucleus of Ki67-negative cells, an effect which was shown to be reversible. Importantly, the proinflammatory cytokine cocktail suppressed Ago2-regulated cell cycle control genes, and siRNA-mediated depletion of Ago2 interfered with cytokine-induced growth inhibition. Conclusion: IFN-γ and TNF induce a stable cell cycle arrest of cancer cells that is accompanied by a fast nuclear Ago2 translocation and repression of Ago2-regulated cell cycle control genes. As Ago2 downregulation impairs cytokine-induced growth regulation, Ago2 may contribute to tissue homeostasis in human cancers.